Power output is a basic technical spec to determine state of AESA capability. If such basic information is not known, how can you reasonably conclude or speculate that the Chinese have made progress to western standards? T/R module count is insufficient because it doesn't give a qualitative dimension.
The power output of a radar is definitely a very important indicator, but range of a radar vs a specific RCS or a specific type of target is also a decent substitute for power output (more often than not, range is the most marketable specification that companies use, and one usually needs to dig a bit deeper to find the power output for a radar).
The problem is that we do not even have reliable range estimates for many more recent Chinese radar systems.
At the end of the day, we simply do not know many of the details or even basic parameters of many Chinese military products which are essential to determining how competitive a product or product type is in comparison to other products even on the more basic levels.
This applies for a wide range of products from radar, to stealth, to sonar, to missiles, datalinks, EO sensors, so on and so forth.
The best we can do is to take information which we know to be true or very likely to be true and play around with certain assumptions and premises to reach an estimate... with full knowledge that the conclusion is only an estimate and will change if new information comes to light.
It's never been about making unassailable conclusions, but about making the most accurate estimates we can despite limited information and evidence, for the purposes of discussion.