Yep, all very good points. Neuropsychopharmacology is extremely and recursively complicated. It's turtles all the way down. Just take one component in all this, the receptor. One might want to think of a receptor as just a switch triggered by a chemical. But autoreceptors are a thing, as you mention. Receptors do different things in different areas of the brain. Receptors form complexes with eachother(e.g 5-HT-2a and D2), and we have essentially no idea why? Receptors have all sorts of different modes of interaction both with ligands and cell internals. Agonism, partial agonism, antagonism, inverse agonist, positive/negative allosteric modulation. The GABA-A receptor is not really one receptor, it's more like a family of receptors made up of a varying constellation of subunit proteins. Different constellations appear in different parts of the brain, and have different allosteric binding sites, which is why benzos vary widely in qualitative effects despite all being "GABA-A PAMs". NMDA receptors don't just bind to glutamate, but also glycine, and magnesium needs to be around? And NMDA triggers an intracellular cascade that regulates membrane expression of AMPA receptors. This is thought to be involved in memory. Cannabinoid receptors are expressed presynaptically carry signals in reverse, so that's weird. And there are 5 identified cannabinoid receptors. We only sort of understand what 2 of them even do. The other 3 are still quite mysterious(last time I checked). Most of the well studied neurotransmitters have not one, but many different receptors they interact with. 5-6 dopamine receptors, I can't even remember all the serotonin receptors, etc. Many of them are still poorly understood.

And of course, neurotransmitter systems talk to eachother. Serotonin so much so that it's also been called a neuromodulator. Because it very often regulates release of other neurotransmitters(including itself).

It's a field that, the more I dig into it, the more confused I get, honestly.

So when I see someone say herp derp, serotonin hypothesis is false, therefore SSRIs are ineffective, the only conclusion I can make is they haven't even tried to dig into it.