Fwiw, engines can run richer or leaner at different rpm, and there is the soft transition from increased velocity versus wind resistance. On modern cars with low CoD, the ideal speed is actually pretty high so getting better mpg at ~65 than ~55 is believable. Once you get to a certain wind resistance mpg will start to drop as the distance travelled requires more work (and fuel) to maintain.
In general, most fuel is consumed getting that 4,000 lbs of car to move in the first place - changing inertia requires a lot of energy, but once the vehicle is moving, very little is needed to keep it moving, you just need to overcome friction.
This is not entirely correct. Wind resistance increases as the square of the velocity, so the drag increases significantly from 55 mph to 65 mph (about 40%); there is no practical situation where the drag profile is more favorable at a higher speed than at a lower one.
I think what you're trying to say is that low-speed operation of a vehicle is inefficient because all of the road friction, mechanical drag, pumping losses, etc., are relatively large in proportion to the wind resistance. The car is using a certain amount of fuel just idling, and cannot go below that threshold, while it is traveling very little distance. However, there is some optimal speed at which the distance traveled is greatest in proportion to the amount of fuel above idle that is consumed. This is typically around 35 mph, after which the drag becomes a significant factor. And don't forget that rolling resistance for the tires also goes up with velocity.
This is why modern automatic transmissions with very high overdrive ratios perform so well with regard to fuel mileage; the engine is operating barely above idle at freeway speeds. It is also operating in a relatively efficient part of it's band...larger throttle opening to reduce pumping losses instead of light throttle at higher rpm. A full-sized car requires about 15 hp to maintain 60 mph (you're right, it takes a whole lot more to get it there quickly). Consider that the accessory drives on the engine probably often consume more than that, so the total fuel consumed would increase by only 50% if we doubled the amount of power we put onto the road.
In any case, as the vehicle speed increases, the idling costs become relatively smaller and smaller, but the drag increases fast; fuel consumption goes up proportionately, so it should get progressively worse for every mph you travel above about 35 mph-or-so. My Acura follows this behavior very closely: It gets phenomenal mileage at 55, slightly worse mileage at 70, and it starts to fall off noticeably at about 85, which I actually do have the ability to cruise at for long distances. Under those same conditions (steady cruise, not start-and-stop), the car would actually do much better at 35 mph, which is why the drivers tried to keep their speeds down in the old Mobilgas economy runs (what we'd call "hypermiling" today).
So, I'm at a loss to understand how the car could get much better mileage at 65 than 55...especially considering that it is in 6th gear in both cases. The only thing I can theorize is that there is something about the variable valve timing or fuel injection programming that happens to be in an unfavorable position around the lower speed. Or that the conditions really aren't equal, and something about the duty cycle is different under the two different speeds.
Interesting point, though...if I buy one of the cars, I'll have to watch for this.