I would still score civ 6 a B+/A- on city state variety. The key reason being that there is a strong mix of city states I read the name of and go "That absolutely should be in" and times I go "Where is that place. I will have to do research". That is a great sign that the development team is trying.
Civ 6 has been very successful at portraying a variety of less prominent historical players. Partially due to how city states are designed around suzerain bonuses now. That way they are not just names, but core parts of strategy that teach the player narratives about why the city state was signifiant.
As far as variety, there are ancient entities without a decipherable record (La Venta, Mohenjo Daro), mid size players that could work as independent civs but would get passed over for immediate neighbors (Preslav, Kabul), and even modern-day city-states (Hong Kong and Vatican).
The big diversity criticism comes from regions where there a lot of city-states on top of densely packed civs, with western Europe being the worst offender, and Cardiff standing out as the most superfluous of all city states. Not every British entity needed representation over other regions of the world. Another region getting close to packed in the Indian subcontinent, with 4 city states but they get more of a pass given how diverse it is. But there are still key gaps.
While Latin America has been wonderfully filled, Cahokia being the only non-replaced city state for both the U.S. and Canada stands out like a sore thumb. I can easily list off a full set of prominent Native American groups that could have unique city state bonuses. Onodaga, Tuskegee, Werowocomoco, Chaco Canyon, Moson Khani, Masset.
The Caribbean is a small enough region geographically that I am not surprised it waited until NFP to get a city state, but more surprise it represented the Taino. I would say if a second Caribbean city state was added my order would be Port-du-Prince over Havanna or an Arawak or Carib settlement, because even if I am on the fence of Haiti getting a civ, it absolutely deserves a city state.
East Africa needs something else, be it Mogadishu, Mombasa, Nairobi, Kampala, or Nyanza. Southern Africa could use an entity from Zimbabwe, with Mapungubwe, Zvongombe or Khami being possible fill ins since Great Zimbabwe is a wonder.
On the other hand, West Africa is fairly well covered with city states, but I would still be up for the Ashanti getting upgraded to a proper civ and the bonus moving to another polity, probably in Cote D'Ivoire.
Central Asia can absolutely fit another entity in to go along with Samarkand, even if Chinguetti's bonus was frustratingly perfect for Khotan.
The big elephant in the room is the vacant space in China. The ability to represent Tibet, Taiwan, Uyghurs, Manchu and other groups would be nice, but the only city state would even have a chance not cause a snafu would be a funky Macau trade city state that gives gold for entertainment complexes. That's a nope for me.
Russia however, has far less resistance to such a problem. There are enough archeological sites of steppe nomads and other settlements past the Urals that would be a pain for a civ to play as on TSL, but can be added to spice up the area. That said I will roll my eyes at Tuva.
For the pacific, I would enjoy something that is not cultural. Pufferfish helmet warriors, Yap stones as another gold generating tile improvement, faith from volcanic eruptions, there are a lot of possibilities that I would love to be surprised with.
