Just going to throw my 2 cents in:
One of the biggest problems with fuel efficiency in an IC engine is the design of the engine itself; this design hasn't changed much in about a hundred years. It is still more-or-less a piston sliding up and down a cylinder, connected via a crank to a crankshaft. Along with a camshaft opening and closing valves, connected to the crankshaft via some belt/gear/chain and pulley arrangement.
In other words, a lot of parts - a lot of power (and fuel) robbing parts. Reduce the part count and/or the connections between them, and you'll gain a more efficient engine (in theory). Think about the following:
1) Why are the valves actuated by the engine? Why not have them electronically/electro-mechanically controlled?
2) The piston slides up and down - but due to side forces from the crank action, there is side-to-side rubbing and wear (causing cylinders to take on an elliptical shape over time). Eliminate that extra friction!
3) Why a crank at all?
4) Note that the piston has to be accelerated down, then brought to a "halt" and driven back up (other pistons firing, plus the flywheel to even things out); this back and forth motion, while balanced, isn't as stable as say something like a turbine, which just spins (but has its own issues - such as needing to run at a fixed speed for maximum efficiency).
5) Turbine hybrid vehicle, perhaps - where the turbine runs at a fixed speed, of course...
All of these issues and problems have been explored, and engines invented/researched to address some or all of them. The Wankel and other rotary engines attempted to solve some of these issues. I've read about one engine that had double-ended pistons, where one end of the piston drove hydraulic fluid through turbines (eliminating the crank and crankshaft). Not too many years back, there was the McMaster motor (which was a nutating disk engine).
Many of these ideas and such have had issues which may have "doomed" them in their time (but maybe today, with today's new materials - something better could be done?), but I sometimes wonder if some of them just died for lack of funding, or if they were bought by "interests" and are just being sat on, waiting for the day when they are absolutely needed - and sold to the public for $$$ as a "new invention" or something. I mean, many of these engines and ideas originally came about in the 1960's or earlier - then seemingly nothing more about them. The McMaster was one of the latest; it seemed like a workable idea, but the website went down a long while ago - I don't know what happened with it.
Speculation and conspiracy, I guess...