Using HST as an example, staring time is key because (a) it can do it, and (b) it must do it to compensate for the small 2.4 meter apeture.

By comparison the OWL telescope (if built) could capture as much light in 3.5 minutes as the Hubble Ultra Deep Field did in 11 days. The extreme HDF exposure times achieve the desired image, but severely limits the cumulative scientific payload -- it's out of commission for days building that one image, whereas OWL could do many such images per night.

The larger diameters available for terrestrial telescopes reduces the extreme exposure times the smaller space telescopes are forced to use for faint objects.

I mention this not to knock Hubble but in answer to the original post lamenting that the reduction of JWST visible spectrum capability would mean no further resolution improvments beyond HST. It's likely those improvements will eventually happen, only from ground-based instruments, and possibly in the same timeframe as JWST.

Terrestrial telescopes (even in optimal sites) might have a limiting magnitude imposed by background atmospheric "sky glow", but I don't know what that is.

And obviously spectral regions blocked by the atmosphere can only be imaged from space.