Water Towns, Mining Towns, Railroad Towns, & Spring Towns

As a kid and even as a young adult, I never gave much thought to how the towns of the Ozarks that I was familiar with got started. I more or less assumed that people just got together and decided they were going to lay out a town, which I guess is somewhat correct as far as it goes, but there was, of course, more to it than that. With few exceptions, there was a reason why towns were located where they were. Rarely did people just get together and decide to create a town in the middle of nowhere.
Offhand, I can think of four reasons why the towns of the Ozarks were located where they were. In other words, we might say that there are four historical categories of towns. I'm sure there may be other categories of towns or other reasons for why they were situated where they were, but these are the four categories I can think of without giving the subject a lot of thought:
Most, if not all, of the older towns of the Ozarks (generally speaking, those that predated the Civil War) were located where they were because of their proximity to a source of water--a river, stream, or spring. A few county seats might have been situated where they were merely because the site was near the center of the proposed county, but even most county seats that I'm familiar with are relatively near a source of flowing water. Lebanon, county seat of Laclede County, and Marshfield, county seat of Webster, might be exceptions. I don't know of significant streams near Lebanon and Marshfield, but maybe there are such streams. Or maybe they were situtated where they were because they were along the old road from St. Louis to southwest Missouri. See, I'm already coming up with other possible categories, but I digress.
Another group of towns got their start as mining towns, notably a good number in the extreme southwest part of Missouri, like Granby, Joplin, and Webb City.
A third group are what might be called railroad towns. Monett is a good example. Originally called Plymouth or Plymouth Junction, it got its start in the late 1880s as station along the Frisco Railroad. Other towns may have existed as little more than wide places in the road before the railroad reached them but didn't really start growing until after the railroad came. Ash Grove is an example. I would still consider towns like Ash Grove "railroad towns."
A fourth category would be mineral water towns that sprang up, most in the 1880s, during the mineral water craze that swept the country. In a way, these towns might be considered water towns, too, but the water was a specific type--mineral water from springs sought for its supposed medicinal properties as opposed to water for merely quenching thirst or streams for navigating boats on. So, I consider these towns a separate category, and I've talked about them on this blog before. The most noted example is Eureka Springs, but there are numerous others.
Are there other categories I've omitted? Do towns situated near a main thoroughfare, such as the possibility of Lebanon and Marshfield I mentioned above, constitute a fifth category? How about county seats located where they were simply because the site was near the center of the county?