I used to be of the mindset that I would get a smaller battery because I don't need anything larger 95% of the time, and can rent if I do.
My mindset is changing though, and at the price points you mention, I would definitely go with the 90kWh pack, although I really think that price point is not realistic for another 3-4 years, but even so, I would probably pay $45K for a 90kWh pack, under the right set of circumstances. Here's my thought process:
Maybe it's my early LEAF, but between battery degradation and poor performance in winter (and I'm talking about North Carolina here, not Minnesota or northern New England), it's definitely a struggle for me in the winter. If I didn't have charging at work, I'd basically be screwed at this point 4 years in. My wife's 2013 seemed more resilient in the winter, but I just think I'd want the additional buffer so I can have confidence the car will be useful in 8-10 years. So I'm already up to the 60kWh pack at a bare minimum.
Second consideration: we would like to take "regional" trips (defined as 200 mile each way) a few times a year. I'd like to be able to do this with one QC stop, and "a few times" is defined as "enough that renting a car for each of these trips would be a hassle/barrier to taking the trip". The 60kWh pack would probably be sufficient to meet this need when the car is new, but I'm unsure about winter when the car is 5+ years old, and given that these are mostly highway miles. It's starting to stretch into the 90kWh pack.
Third consideration: we take an annual trip that's about 750 miles each way. Renting is an option here, but would still be nice to not have to do this. For this option, I really feel that you need a 350 mile range EV, the argument being that at 80% per quick charge, you're basically talking about 280 miles per QC, or about 4 hours of driving. Given that we try to make this trip in one (albeit long) day, this is really the minimum range we'd need to make doing this trip realistic. Spacing the stops out and assuming starting with a 100% charge and 80% after each stop, this trip is fairly comfortable with 2 stops in a 350 mile EV (stops at around miles 290, 520 and 750). Could we do it in the 300 mile EV? Technically yes (miles 290, 520 and 750), but this only leaves about 10 miles of buffer to account for highway speeds, cold weather, battery degradation and non-ideal charging station spacing). But adding a third stop would not be the end of the world for this trip. Certainly we'd be stopping for a rest room break anyway, so working in a slightly extended stop as a third stop would probably work fine. A 250 mile EV would probably not work for this trip though, and certainly not a 200 mile EV like the BoltEV.
The caveat here is that there would have to be charging stations more or less ideally spaced along the route. Most of this route is on I-81, and between Harrisburg, PA and Watertown, NY, there doesn't appear to be a single CHAdeMO station. Tesla Superchargers exist in Binghamton, NY and Syracuse, NY at least, but nothing in PA on I-81. Hagerstown, MD to Binghamton, NY is 254 miles (cutting it pretty close!) Of course I expect this to change between now and the time that these vehicles come to market, but that will be a consideration. There is no point in going to the 90kWh car for me unless there is sufficient charging along this particular route. A Tesla Model 3, assuming compatibility with CHAdeMO through an adapter, would meet this need today. Hopefully Nissan is working with Tesla on Supercharger access for the LEAF 2.
The final consideration is this: I certainly would not need 2 vehicles sized like this. One for the family is sufficient, the other one can be a 100 mile class vehicle. However, if V2G ever takes off, and at the right price point, I can see the benefit of having a larger battery than you need on a daily basis. I would love to be able to contribute the 50% or so of my battery I am not using to leveling renewable resources. This line of thinking is probably the exception--most people would only be concerned with their bottom line cost.