The same Israeli study shows that infection-mediated immunity is waning: if they don't line up infection times and vaccination times and allow earlier infections, the protection drops to 6x from 13x. And that includes the recent cases, so the old cases are worse yet.

That being infected doesn't count in the U.S. the same as having had a vaccine is ridiculous--in Europe it does, and places with a "vaccine passport" allow it as an alternative.

But it's also not good to overplay the protection you get from having been infected. Protection does go down, but it starts high (at least as high as the mRNA vaccines) and drops more slowly, apparently. (We're still waiting on good publications that plot the timecourse, though you can find evidence of it in Oklahoma's COVID-19 reports.)

If you're going to ask people to take a booster for the vaccines, it's quite sensible to ask them to take a booster after infection too (just maybe not as soon).

Boosters (of the Pfizer vaccine) work incredibly well, with 11x protection beyond what the (waning) vaccine will do: https://www.nejm.org/doi/full/10.1056/NEJMoa2114255

So, anyway--getting immunity via infection should certainly not be discounted (especially compared to less-effective vaccines like J&J's!). But it doesn't follow that a booster is unnecessary. (There is no evidence I'm aware of that a *full* dose is necessary (i.e. two shots of vaccine after recovery from COVID-19 instead of just one as a booster).)