Be sure to understand that the "speed" reported by networking equipment (and specifications) as "link rate" is not the same thing you see when running a "speed test." Speedtest does not test the "speed" (link rate) of anything, speedtest sends out a measured amount of data, times it, then computes a statistical average - like the trip computer in a car. Don't fall into the trap of thinking that link rates and speedtest results are "the same thing" just because they are both reported in the metric of "bits per second."
For example, if you hook up something via (say) gigabit ethernet (which works at a fixed rate) it defintely is a "1000mbps" link rate, no matter what speedtest reports because ethernet does not work any other way.
Wi-Fi isn't really "spewed out" by anything - it would be a good idea to ditch from your mindset any notion of Wi-Fi as some ethereal energy field like The Force or Ley Lines generated by some magical "Wi-Fi generator." Wi-Fi is a two way radio "conversation" like walkie-talkies not a one way radio "lecture" like television. Think of it in terms of sound and I think you'll get the idea.
To repeat, the link rate (speed) and throughput you see reported connected to you router using ethernet is completely unrelated to what it offers over Wi-Fi. One cannot infer the "because the ethernet rate is X it should be that good over Wi-Fi" - it doesn't work like that.
Not least because the throughput (as measure by speedtest) for ethernet is not directly comparable to Wi-Fi. A percentage of the basic link rate is "lost" to things like error correction, management chatter, encryption and so forth. There's a rough "wet finger" metric called the "protocol efficiency" that give some ball park measures of such losses. Ethernet has about 97% protocol efficiency, Wi-Fi is less good at of the order of 55-75% (on a good day with ideal signalling conditions.) Add some interference, distance, airtime competition and so on and the protocol efficiency of Wi-Fi drops as the error rates go up, retries increase, more conservative link rates and modulation schemes are used. It's all very complicated!
By connecting up via gigabit ethernet and speed testing, you've effectively confirmed that your ISP link is performing (as it's so much slower than gig ethernet) and your router has the horsepower to "route" the required amount of data. So you've essentially localised this to a Wi-Fi issue, either in the router, the clients, the signalling conditions or a combination of all three.
Unfortunately statements on your iPhone spec like "does Wi-Fi 6" are pretty much useless. We need to know things like the number of spacial streams (it does mention 2x2 MIMO,) the channel bandwidth supported and a whole bunch of other stuff to come up with useful numbers. Apple have always been pretty crap about providing such information.