Photorealism is very doable in VR, and has been done in several cases. Just Valves environments alone are believably realistic, from the home-zones to The Lab demo (where you stand on the mountain and play with your robot dog

), and there's an app where someone did 3D photocapture and inserted them in "game" (more like an art gallery.) Photorealism isn't holding VR back, it's just expensive to do well.
As far as MMO's go, OrbusVR is pretty solid, but it's definitely indie, and closer to a runescape-meets-wow than anything.
Elite Dangerous in VR is phenomenal.
Also, there's no reason why there couldn't be device-asynchronous communication between a VR person and a webcam person. It's really just software support. Of course, the VR person would be represented as an avatar, and the webcam person could be handled multiple ways - either via an "in VR" flat-display, or it could translate them into a VR avatar itself.
I don't know what you
@Rebis means by calibrating or response times.
@peoplesuck:
https://en.wikipedia.org/wiki/Neuralink might interest you - elon musk is pretty much trying to build out something like what you're asking about. His particular application is more about directly interfacing with a computer, not having the whole display rendered mentally, but I imagine that technology will evolve from something like neuralink.