In general I think the balance of early game super disasters being able to drop 6-10 fertility on you while simultaneously mostly affecting empty tiles (since you haven't had much time to build up yet) is pretty good.
They're all avoidable except for edge cases of storms that cross a biome boundary. Although floods drop so much fertility the boosts to tiles usually outweigh having a farm or mine in the first place. As an aside, volcanic soil allows some odd placement -
any volcanic soil can hold a farm, mine, terrace, etc, regardless of hills, snow, etc or your tech level. (Very handy!)
Storms, as in Dust or Blizzard. Have actually not had that much trouble from them. When I play on a lot of blizzard-prone Tundra, I'm usually playing Russia, so I have my Magic Petrine Anti-Blizzard Cream to at least protect my units. In the desert, I admit that I assume anybody crazy enough to go there deserves what they get (having lived in a few deserts in my time)
...
There is, simply, no single Disaster or set of disasters that will completely stop you in the late game.
In my experience, the game has a tendency to select a few specific spots in the world deserts for dust storms. The problem is that this means if you play a civ like Mali that is extremely desert centric, then you may have nary a mote blow into your bazaar. Or you might get pounded by chain Haboobs continuously. It's a distinctly late game issue because the frequency of storms gets cranked up with CO2 levels and if you own and have developed (usually running something like desert folkore, so it's a bunch of HS/Sugubas) the doomed desert, it's just awful. I have had swaths of 4-5 cities get hit by a haboob, and then before the first one is done, another haboob forms in the same desert. It's maddening. I should have taken screenshots; the entire screen is burned out districts and sand encrusted mines. Even on disaster setting 4 it's completely out of character with the other disasters.
How will I peddle trinkets in my trade posts with all this sand?!
Alas, to change all that would require a rework of too many major elements of the game, to reverse a trend that has now been in effect for two iterations of the game (Civ V and VI). I don't see it happening - much as I might like it to.
Well, civ6 cities can grow
much faster than civ5 cities ever could. I can put up a city in the industrial and have it size 20 before the game ends. In civ5 you basically knew that after a certain era you would
never get any new city above a certain size. The utility of military engineers to rush out infrastructure that usually must be hardbuilt, and reyna/moksha being able to buy, closes a lot of the gap in development. Getting a city to size 10 fast is a matter of a few farms under replaceable parts. But if you don't use the governors then building districts, even in otherwise productive new cities, is a total chore. It might be better in the future to have a scaling district piece based on the number of specialty districts in that city already, and compound that with the game time scaling, so that new cities can still get on their feet quick. And it's not like you can't develop new cities in the late game; it is just extremely demoralizing to see a 120 turn build time on an aqueduct so we don't like doing it.