The tech to keep your eye on is orbital refueling. If someone can finally crack that, all bets of this sort are off. While I'd love to see a nuclear rocket or something, even just the energy advantage of being able to refuel in orbit rewrites all our intuitions about time and expense of exploring the Solar System. We are so bound by having to launch things as single bundles of stuff from the ground... the rocket equation does not like that.
>> even just the energy advantage of being able to refuel in orbit rewrites all our intuitions about time and expense of exploring the Solar System.
Except that it doesn't. Refueling in LEO only helps if your gas station is in the same orbital inclination you need for your target. As each interplanetary launch will be via a different orbital plane/inclination, there is little use for generic infrastructure. Just using a bigger one-time rocket will be more efficient than trying to refuel and then boost interplanetary from an inappropriate orbital plane.
Now in higher orbits refueling can start to make sense as inclination becomes less of a handicap, but boosting from a higher orbit is less efficient than from a lower (oberth). A single larger rocket will still be the way to go.
If you're launching the fuel for the purpose of extending the range of a probe, well, of course you'll put it in the right place.
There's no point of conceiving of a world where we can afford to launch fuel but still have to treat every single launch as exceedingly precious and expensive such that we can only afford to set up One Fuel Depot in space, the way we have One Space Station. Either this gets cheap enough that SpaceX basically has a "launch + more fuel launch(es)" as a standard package or it never happens at all, there isn't much in between.
