Do we really need much more pixels?
The biggest problem at the moment is not so much resolution as fill factor; the so-called "screen door effect". You can see individual pixels not so much because the resolution is too low, but because you can see the gaps around the individual pixels. Things don't look blocky as when the resolution is too low, but it looks as if you are looking through a fine mesh. Even just using my Pixel and Daydream headset, not the highest resolution rig, I rarely really feel like resolution is what's lacking.
Now you could solve this simply by throwing pixels at it - the higher the resolution, the finer the "mesh", and if you go fine enough, then it will seem to disappear. This will, as noted, require a great deal of processing power to render all those pixels, however, and seems like a pretty brute force way of dealing with it.
Foveated rendering doesn't sound to me like much better of a solution. The complexity of getting the eye tracking to work well enough and fast enough, the variable-resolution rendering, and the other issues mentioned, sounds like an absolute nightmare.
Now I don't know displays, but I do know a bit about image sensors. Image sensors also suffer from fill factor issues - light falling between the pixels is wasted, thus the sensitivity of the sensor is less than optimal. One common solution to that is micro-lenses - as it sounds, an array of tiny lenses is bonded over the pixels, and the lenses capture light that would have fallen into the dead area around the pixel and focus it onto the active area. I don't see any reason why something similar wouldn't work for displays, and a quick bit of Googling says I'm not the first to think that way. No idea of the practicality or cost of manufacturing that kind of thing, but it may well be that we don't need so many pixels to make VR less crap after all, a 4k display plus micro-diffusers (or whatever those things end up being called) to fix the fill factor may be more than good enough.