A few thoughts:
It's true that most older aircraft have 'atrocious' fuel consumption, but that's as much to do with the airframes as the engines. I've heard it said that Rotaxes are not really much more efficient than Lycosaurs and Continentals in the cruise, but that their advantages lie in climbs and circuits... Where admittedly a lot of aircraft spend a lot of time. Particularly training aircraft.
WRT car engines... if you take an engine and run it at 20% of peak capacity, it's not surprising if it lasts for many thousands of hours with little wear. But what if you put it in your aircraft and run it at 100% for minutes at a time, then spend the rest of the time at 75% power?
There are older aircraft that allow you to advance/retard the ignition; presumably it was felt that 4 levers (throttle, carb heat, mixture, pitch) was already enough for any mortal to get their head around.
Certification may not prevent engine development, but aircraft certification prevents you from just bolting a 100hp rotax to anything that is designed for a 100hp continental. Contrariwise, if you were to buy a new aircraft, would you want an engine from an unproven manufacturer who might go bust at any moment - thereby grounding the whole aircraft potentially forever - or would you get an engine from a vendor who's been around for the best part of a century and for which 3rd party manufacturers supply approved parts? If an aircraft cost £10,000 and was expected to last 10 years you might well chance the modern engine with a slightly lower fuel burn, but as a new Cessna costs fifty times that and might be expected to last 50 years, knowing that the powerplant is likely to be supported for decades to come is a big issue.
Personally I fly behind a VW engine which, five decades after it was made, is likely to have parts support for many years to come.