Trouble is the article misses a very important point: that's just antibody (B-cell mediated) immunity. There is also T-cell immunity, which is harder to quantif

Trouble is the article misses a very important point: that's just antibody (B-cell mediated) immunity. There is also T-cell immunity, which is harder to quantify so it doesn't get much attention, but we can be quite sure that it plays important role because many people who recovered from COVID don't develop effective neutralising antibodies at all. To recover from a disease like this you have to develop an effective immune response, so if it's not antibodies, it has to be T-cell immunity. Moreover, even for antibody immunity it's not the whole story, because it's normal and expected for antibody count to decline with time — we have memory cells that ensure rapid production of the relevant antibody when the same antigene is presented. It's basically the case of paying attention to something we can measure just because we can measure it.