Additional cons:
- Increased noise from the regeneration
- Potential for distortion introducted by the regeneration
- Non-flattopped response because of the single resonator
(so you need to make the resonance broad enough that
you can do a decent job with another filter, likely at a lower
frequency)
- To get high Q with reasonable regeneration still requires a
BIG resonator (at VHF). High regeneration is difficult to
make stable.

I'd enjoy reading about the results of your experiments with this, but
I honestly don't have high expectations for its being practical.

A long time ago, I worked on a receiver that applied gated regeneration
to the energy introduced into a cavity by an antenna, up in the several
GHz region, and it was a sensitive design for a very specific purpose
that worked well with few parts. But that design was optimized to
accomplish a special task, and the design approach was appropriate for
that task.

Cheers,
Tom