(FORTUNE Magazine) – The Detroit auto show always glitters with gleaming cars and spectacular visual effects. This year, though, there was a change in the air, and it wasn't from gas fumes. Sure, the annual gala filled the show floor with black-tied executives and their spouses. Of course there was the usual mix of one-upmanship and bonhomie. But there was also an almost tangible sense of renewed purpose. The CEOs of America's Big Three (Rick Wagoner of GM, Bill Ford of Ford, and Dieter Zetsche of DaimlerChrysler) even demonstrated their confidence by going on stage to sweep the drapes off their new designs--precisely the kind of public display they have often avoided. The implied message: Detroit is pulling out all the stops.

That isn't just a matter of cupholders and crushed-velour upholstery. The three companies will unveil a record 48 new or revamped sedans, coupes, minivans, sports cars, sport utilities, crossovers, and pickups in 2004. Look under Detroit's hood, and the change is also impressive. By overhauling the way they design and make cars, the Three have cut product development costs sharply. That allows them to get new models on the road faster and cheaper than ever before. "We're building great products--the best we've ever made," Wagoner told Wall Street analysts in January.

It's about time. In 2003, Detroit fumbled away another 1.5 percentage points of the domestic market to foreign competitors. It now accounts for only 60.2% of vehicles sold in the U.S.--the lowest percentage ever. Of course there is nothing new here. Detroit's market share started declining when the Beatles were still together and has been falling ever since. Pessimists see the U.S.-based automakers spiraling toward irrelevance, selling cars that don't match those of their international rivals in value, quality, or sex appeal.

Toyota executive vice president Jim Press reflected this feeling in January when he suggested to analysts that the Big Three should be renamed the Detroit Three to better reflect their shrunken global significance. Even longtime industry supporters like David Cole, chairman of the Center for Automotive Research in Ann Arbor, is bearish on Detroit. "I think the market-share erosion is going to continue, if for no other reason than the international guys are in more [product] segments," he says.

Save the obits. Detroit is mounting a vigorous comeback. The Three believe that their new lineups, which provide a larger variety of models at a broad range of prices, will stop the bleeding and get them on the road to recovery. GM and Chrysler expect to gain market share in 2004. Ford's leaders are confident that they will hold steady despite fewer sales to rental-car fleets. Says Chrysler's Zetsche: "2004 is no fluke. We're going to continue to launch new and exciting vehicles at a fast pace--and attack virtually every segment of the auto market."

Zetsche made his point in spades at the auto show when Chrysler revealed an over-the-top 850-horsepower supercar that is designed to go faster than even Mercedes' speediest racer. Not to be outdone, Bill Ford took the wraps off the production version of Ford's 200-mile-per-hour GT, while Rick Wagoner gushed over the new 400-horsepower Corvette. "The formula for success is pretty simple," Wagoner says: "a combination of great products, along with continued excellent quality and cost performance."

Is that the usual bluster from the usual suspects? Perhaps, but this year they are not the only ones who think the industry is turning a corner. Detroit economist Jeff Schuster of J.D. Power & Associates sees share declining for the next two years--but at a much slower pace--before stabilizing. Dave Andrea, who watches the industry for the Original Equipment Suppliers Association, thinks the Big Three will hold the fort this year, thanks to new models and aggressive pricing.

Give the Motown guys credit for finally realizing that warmed-over product concepts and me-too designs won't cut it. Detroit's reputation among consumers for spitting out dull clunkers won't be reversed in a year, particularly since the Big Five imports--Toyota, Honda, Nissan, Mercedes, and BMW--have built strong brands and enormous momentum. But at least Detroit is putting itself in position to compete. In the past few years the Three have closed plants, sloughed off stale product lines like Oldsmobile and Plymouth, and streamlined operations. "It is fourth down with no minutes left to play, so [Detroit] better pull something out of its helmet." says Karl Brauer, editor-in-chief of Edmunds.com, an online auto information service.

What Detroit has pulled out is those 48 new models. A combination of competition, capability, and common sense has allowed U.S. automakers to cut the costs required to design and engineer new cars. The old rule of thumb was that it took four years and $1 billion to get a car from the drawing board to the road. No more. Executive vice president James Padilla says that Ford has reduced its product-development time by 25% since the late 1990s and is getting better by another 10% a year. Ford produces almost a third of its new models in 30 months or less, according to product-creation chief Phil Martens. That's still behind the best of the Japanese, but the gap is narrowing, which is one reason to believe that Ford can stabilize its place in the market.

The poster car for the industry's modern product development is a sweet little sports car, the Pontiac Solstice, which goes on sale in late summer 2005. Under the old economics the Solstice would never have made it into production. Prospective returns on the two-seater were low--20,000 units a year, with a base sticker price under $20,000--and its costs were high. So if the car was ever going to get off the drawing board, the costs had to crash.

The biggest savings were made in the factory. To start, GM cut the cost of the dies used to stamp the body parts by $50 million by using computers to design and validate them and by using an economical new technique called hydroforming, which eliminates half of the die and replaces it with a water-filled bladder. GM saved another chunk of money--up to $80 million--because the Solstice has an uncomplicated body that can take advantage of existing tooling. Even those savings wouldn't have been enough to put the Solstice on the road. So GM based the car on a simple architecture called Kappa that it can use as the foundation for other small rear-drive cars. Adding it all up, the Solstice goes into production for $100 million, and GM has a spiffy little roadster that just might help make Pontiac cool again.

While GM is getting a big publicity burst, Chrysler spent four times as much money to make improvements, mostly hidden from sight, to a pair of minivans. Sales of the popular and highly profitable Chrysler Town & Country and Dodge Caravan have been under siege from Toyota and Honda, which make vans that come equipped with third-row seats that can be folded flat into the floor. Chrysler spent $400 million on a quicker-than-usual minivan redesign, performing major surgery on the floor pan and underbody, so that it could create space to fold the second-row seats as well as the third row flat into the floor. An Edmunds.com reviewer said it was "just what the venerable [vans] needed."

Another way Detroit is trimming costs while improving flexibility is by using the same parts across models. If that sounds like an obvious thing to do, it wasn't on Planet Detroit. There, automakers used to pride themselves on reinventing the wheel with every model. In 1999, when former Chrysler CEO Bob Eaton introduced the Jeep Grand Cherokee, he proudly held up a bag that he said contained all of the carryover parts. His point was that this Jeep was practically a whole new animal. But does it make sense to replace every part with every update? Not everyone thinks so. Toyota and other manufacturers have long churned out new models on the foundations of old ones, which saves time and money and makes shifting production easier.

Detroit's tendency to start from scratch was exacerbated by the way the companies organized themselves. Chrysler engineers worked in separate teams that didn't share enough knowledge or parts. Product-development chief Eric Ridenour says the company has addressed that problem, in the process improving its investment efficiency 30% in five years as well as improving quality. At Ford, meanwhile, engineers working on Ford products didn't talk much to those working on Mercury or Lincoln vehicles or vice versa. That led to expensive redundancies. Ford president Nick Scheele expressed shock when he told reporters in January that the Lincoln Aviator, which is structurally identical to the Ford Explorer, uses different door latches and engine mounts--multiplying the cost of designing, procuring, and testing the parts. Scheele can't do anything about it because the Aviator is being totally changed for model year 2006. Engineers on similar models now work together, and Ford aims to use 40% to 60% of the same parts during a redesign.

Designers are also getting smarter. The emphasis is on creating cars that look different though built from the same basic architecture. GM generates global economies of scale by using its Epsilon framework for the Chevy Malibu Maxx in the U.S. as well as for the Swedish Saab 9-3, European Opels and Vauxhalls, Australian Holdens, and South American Chevys. Ford is trying a different approach by using the components from the XC90 sport utility/station wagon made by Volvo (which it owns) as the underpinning of the Ford Five Hundred sedan and Freestyle crossover wagon. Since Volvos are engineered to sell for higher prices than Fords, engineers reduced costs by substituting stamped steel parts for cast aluminum and magnesium. As for Chrysler, it is borrowing parts from Mercedes, its corporate sibling. The 2005 Chrysler 300C sedan uses a transmission and rear-suspension architecture designed in Germany.

The point of all this is to cut production costs. American manufacturers are racing to catch up to Toyota and Honda by adapting factories for flexible manufacturing. Flexible factories can run five or more different kinds of cars down an assembly line; that helps balance production between fast-and slow-selling models. Flexible production lines are also cheaper to equip and to retool because welders and paint sprayers can be reprogrammed instead of replaced.

General Motors leads the domestic industry in flexible manufacturing and is also reaping the benefits that come from integrating its global engineering and manufacturing operations into one organization. Ford is still recovering from several ill-fated attempts by former CEOs Alex Trotman and Jac Nasser to reorganize product development, and has set off in yet another direction behind product boss Martens. Chrysler dithered for a year or two after the merger. It has since coordinated its new development activities with Mercedes and an-other corporate partner, Mitsubishi.

What all this reflects is that Motown has heeded the example of engineering-dominated companies like Honda and BMW, which have boosted sales and profit margins by offering products that customers consider exceptional. Their success has reinforced the message that products that are viewed as average get sold like commodities; prices drop until the market clears. Profits are thin. That's the territory that Detroit has been treading in and desperately wants to leave behind.

With the new models and the economic wind at their back, the Big Three should be able to make gains at the expense of second-tier European and Asian carmakers such as Mitsubishi and Mazda, and shrink the losses to the Big Five. They are in a particularly good position to make up lost ground, since the fall of the dollar means that Japanese and German imports will cost more.

GM is poised to launch a record 27 models in 2004. Of those, 12 are either new or substantially redesigned, while the others have received sheet-metal changes or a different body style or powertrain.

Ford has reworked its lineup of passenger cars to reverse a striking decline in vehicle market share, from 20.2% in 2002 to 19.5% in 2003. It will add, among others, the spacious Five Hundred sedan; a van--station wagon crossover, called the Freestyle; and a new version of the iconic Mustang. "It's the biggest wave of new product in our history," said Bill Ford during the show in Detroit. "There is no question we are going to have more arrows in our quiver to go after share." But the new models arrive late in the year, and incentives on older ones could leave Ford vulnerable to missing its revenue targets. UBS analyst Saul Rubin described Ford's presentation to analysts earlier this year as "a lot of hype." But someone out there believes in it: Looking ahead to an improving economy, investors have been bidding up the share price. Since reaching a low of $6.58 last spring, Ford shares have more than doubled.

Chrysler has struggled in the aftermath of its 1998 merger with Germany's DaimlerBenz. It ended 2003 with a market share of 12.8% for its Dodge, Chrysler, and Jeep brands, down from 13.1% in 2002. This year Chrysler expects to launch nine new products or major redesigns that, had they all gone into production on Jan. 1, would account for 60% of the 2.1 million vehicles it will produce. These include reworkings of its popular minivans, a new Jeep Grand Cherokee and Dodge Dakota pickup, and a pair of all-new rear-drive full-sized cars: the Chrysler 300C and Dodge Magnum. Says Zetsche: "We've been on a three-year transformation to reduce fixed costs by 30%, and our quality, as measured by warranty costs, is up 30%. Now we go on the offensive." Over the next three years Chrysler will roll out 25 new models.

The bill for all this is astronomical--$18.5 billion in capital improvements for 2002. All three companies are spending more than they can depreciate, which hurts cash flow and weakens balance sheets. Essentially, they are all making the same high-stakes bet: that by investing bigtime in new and better cars, they will score a big payoff. If customers like the cars, the companies can charge more for them and stop eroding margins through no-interest deals and other discounts. That will boost profits, make cash flow in the right direction, shore up their equity positions, restore resale value, and squelch talk of the Detroit Three.

Detroit has botched similar opportunities before. In the mid-1990s, for example, the yen strengthened and customers rushed to buy made-in-America minivans and sport utilities. These big vehicles made big money, and the Big Three cashed in. In 1997 they reported combined profits of $16.4 billion. But instead of reinvesting the loot in successful new products, the companies spent wildly on overseas acquisitions--GM bought a stake in Fiat, while Ford took over Land Rover--that have turned out to be money-losers. Their competitors, in the meantime, stayed focused on keeping consumers happy. Last year Toyota passed Chevrolet and Ford to become the bestselling passenger-car brand in America, while Mercedes and BMW now excite the aspirational urges of the rich and hip in a way that Cadillac and Lincoln haven't for decades. The Koreans are coming on strong. Hyundai now outsells Buick, while Kia has sped past Mercury. By nearly every quantitative benchmark, such as defects per hundred, assembly hours per car, or time to market, Detroit still lags behind the best foreign manufacturers.

So it could certainly fail this time around too, as recent duds like the Pontiac Aztek, Ford Thunderbird, and Chrysler Pacifica prove. Customer appeal, pizzazz, whatever you want to call it--isn't entirely rational. That's why private-school children in New York City get dropped off by mothers driving massive off-road vehicles. At least this time Detroit has the right strategy: to make great cars. With its new models, U.S. automakers are reflecting a passion for winning customers they haven't shown in years.

Fasten your seatbelts: The Big Three won't have many opportunities as promising as this one to reverse four decades of mistakes.

FEEDBACK ataylor@fortunemail.com