I've been thinking about what my Traveller universe might end up looking like.
I'm almost certain to use Paul Gazis's Eight Worlds rules for star ships (and perhaps more of his rules), though I might adjust some numbers depending on what assumptions I make for my universe.
One result of this is that ships aren't constrained to discrete jumps, nor are all jumps one week in duration. A jump 1 ship could take two weeks to travel as far as a jump 2 ship travels in one week (we don't yet know how the fuel costs change). So, rather than the universe being a hex map, each star will have coordinates to some precision (1 ly? .1 ly? .01 ly?). I'm still debating 2-d versus 3-d, and if I use real star locations or not, and if Sol/Earth is part of my universe (probably yes, but still up in the air, is it in the initial locale, or somewhere's distant).
As I have been thinking and processing the various systems for generating stars, and thinking about how to handle interstellar travel, I've come up with a few thoughts:
First, a science fiction universe with interstellar travel will have a bunch of "interesting" stars. These stars are where the main action takes place, or are home worlds, or whatever. These stars need not be occupied by earth-like worlds, but they probably have to have useful resources of some sort.
Now one could have a universe consisting just of those stars, and abstract space travel so those stars are the only ones that matter. Such a universe might even use an abstract map. But such a setup would diminish the importance of ships to the game (and one could even imagine a universe that virtually eliminates the importance of ships, to the point one might as well just have a teleporter network between the worlds of interest).
So then, to make ships and their operations interesting and important, we need to bring in other factors. A simple factor to bring in is to make logistics important. Now it matters how far apart the stars are. Additional stars might become important, not because they have an interesting world to have an adventure on, but because a ship needs to make a stopover for logistical reasons (for example, to re-fuel).
Of course, once such stars gain logistical relevance, they become a place for excitement to happen. During a war, a stopover might be blockaded. Or pirates might infest a stopover. Or some kind of natural event might take place, or there might be a systems failure in the ship.
Now it becomes interesting where you go, and what route you take.
Then the question arises of how many stopover locations are there? Too many, and they become meaningless. For example, say interesting stars are 10, 20, or even 30 light years apart, and ships can refuel by dipping into a star. Introduce hundreds or thousands of stopovers between any two interesting stars (or really even more than a few), and suddenly it becomes unlikely someone would be lying in wait for you. So probably stopovers have to be rare. So we probably can't refuel from a sun. Probably wilderness refueling has to be either slow, or unreliable. Now, we might have a refueling base at a star with nothing much more than a single icy planet where a plant can painstakingly extract hydrogen, perhaps, for a difficult route, just enough for a few small ships per week or two, and they probably charge a lot. But it might be worth it to shave off some time for a rich cargo, or important mission.
Of course, depending on how the logistics work out, tankers might be workable for refueling, but if that becomes too easy, we're back to hundreds of stopovers.
So lots to think about. And none of this particularly depends on what rules are used to generate the "interesting" systems, though they do depend on how one treats the "realistic" data.