Yes.
And "aiming" your observatory involves moving it on that sphere. Given the distances involved that is pretty much either impossible or time prohibitive.
The region starts at 548 AU from the Sun. So 548 times the average Sun-Earth distance.
In an ideal world you would teleport your camera to this location instantaneously, take a picture and then teleport to the next location to look at something else.
We don't know how to do that. The distances are immense.
So instead we pick a target, and send out a satellite or satellites on the opposite vector from it to take a peek. There is a single point where the target will be in perfect focus, but in practice (as the paper shows) the target is "in-focus" enough in a larger region that your satellites can take a picture while they fly through the region around the ideal point.
However, we are using the mental model of a camera lens to reach that conclusion, and I'm not certain it applies:
1) even if we use the mental model of a camera lens, given the distances involved, you can probably consider most of the interesting the targets to be "at infinity"
2) I'm not certain that a gravitational lens works like a camera lens.