'Few' like I can't think of any specificly offhand. I think much less than a literal handfull. Note that Hubble was not designed to be serviced in the manner it's been modified. Quite an accomplishment.

The issue is the recurring cost of a manned service mission vs producing N telescopes or whatever and simply launching ground upgraded improved replacements. I certainly don't know the answer. Sadly, I'm sure NASA really doesn't either.

It just bugs me that the 11 figure Orion development cost delivers a 21st century replacement for the 1970s technology shuttle that has markedly inferior capabilities, excepting long duration mission capability. Not that I'd want to sit in the tiny capsule for weeks on end.