We used LeapMotion on a number of RnD tests and actual projects; while it's not the holy grail of hand tracking, it's a pretty awesome device, especially for the abysmal cost it has.
The point would be, a Leap Motion-esque tracking system, applied to any of the VR headsets, would result in incredible experiences - we strapped one to a DK2 and we had fun times, i can just imagine if a company would make a specific device for that purpose.
Leap Motion already is iterating for VR. And considering it's doing everything with 2 really simple IR cameras ;P
(which is also the reason why it wasn't really included in any major VR headsets from the get go, because all the positional tracking systems use IR, so having a reverse emitter of IR lights to map the hands would easily cause tracking conflicts. could still use different IR frequencies though).
The difference with the HTC Vive Pre's "breakthrough" (lol, market speak, I hoped this 'breakthrough' was coming from Valve's side, instead I discovered was a dud word from HTC's part) is that is a fundamentally different system (and for now, sub-par).
They downgraded from a stereo-camera setup to a mono-camera, they're not doing any distortion, basically it's even LESS accurate a camera than what you're getting using the passthrough function on a GearVR.
As it is now, you can't even attempt to use it for gameplay purposes, since it doesn't correctly reflect the position of the controllers compared to your hand, when bringing the hand closer to the face.
I love the idea of the Chaperone system, but this solution surely is not a "breakthrough". All IMO naturally