We can't really have this discussion unless we define some terms. What we mean by high, low, medium mileage. Also, we need to know how fast our "mileage" is and what distance we're training for.
It seems unlikely, based on our best knowledge, that running at under 60% of vo2max is going to produce any significant aerobic adaptations. Pretty much everyone acknowledges this. Canova says running under 90% of marathon pace is only useful for "regeneration" and getting the weekly total up. Daniels has easy paces that a lot of people find quite fast, definitely 60% or more. Vigil likewise. Coe and Martin. Salazar. And on.
For inexperienced runners, this isn't an issue. They will typically be doing their easy runs 60%, because they are undeveloped and it doesn't take much to stress their aerobic system. But a lot of decently trained runners are doing a significant portion of their easy miles at under 60%. The question for these runners is whether there's any purpose to those miles. I can think of four: Weight maintenance is a major benefit, as the OP noted. It's also likely that the additional light exercise helps to improve your body's hormonal profile (short easy runs produce HGH, contrary to what was once thought). For marathoners specifically, I think that a the repetitive eccentric contractions of high mileage help to produce structural changes that stave off muscle fatigue in the late stages of the race. Finally, for all runners, when you adapt to higher mileage, even if a lot of it's easy, you are then able to drop your mileage a certain percentage and do a higher quantity of quality. That allows people to get more out of their talent. For instance, some top marathoners never go over 13 miles at MP, and others do 20 or more in training. The latter runners have probably done a lot more work to prepare for that kind of effort so that it will actually be a useful stimulus and not just something that destroys them.
I think we have a bit of an issue now where a common, traditional running practice (grouping all runs into either "workouts" or easy mileage) is suddenly getting support from some questionable contemporary research about polarized training, which is causing people to lose sight of the significance of in-between training. People always say that if you run too fast on your "easy days" then you won't be able to perform on your hard days. But that's circular reasoning because you start from the premise that the day was supposed to be easy. What if it isn't?
My approach has been to periodize my moderate-paced running. So early in a training cycle, I run a lot of my non-workout mileage at a pretty solid clip (mid to low sixes for a 2:30 guy, after a few miles of easing into it). This doesn't leave me too beat-up for my workouts, because my workouts are easier at this stage of the cycle. As I get closer to key competitions, the quality days get harder and harder, and in order to be ready for them, it becomes necessary to go easier and easier on other days. I've found that this always brings me to a pretty good peak, but it also means that I end up cannibalizing some of my base, and it's necessary to go back to easier workouts and a lot of moderate running before I can make a similar push again.