There are a lot of jobs that Hubble is good for. The two that he cares about happen to be among them.

To pick another of more interest to you, if we want to some day carry out a comprehensive survey for major asteriods and comets of size over 1 km across that might be on an intersection path with us, then guess what, Hubble or relatives are the only way to do it. (If we know about something of this size a few decades ahead of time, we can probably do something about it.)

In other words adaptive optics is useless for the only astronomical survey that is likely to have defence implications for us.

Furthermore many people believe that it is possible to dramatically lower the cost of operating in space. In which case the economics against space-based telescopes would go away. And the same survey which was estimated as being a security issue becomes the groundwork to identify what space-based objects we might want to try to mine!