Also quite similar to the Buddhist concept of emptiness.

I was sort of describing a stance rather than presenting a stance of my own. I understand your first paragraph here to be something of an elaboration of my last paragraph.

My own stance is, however, remarkably similar to what I laid out. I don't have a particularly detailed understanding of Singer's work--I've only read brief papers by him, and not many of his longer works. However, I would describe my own position briefly, if you're interested.

It is a long tradition in philosophy going back to at least Plato to analogize between the individual (I almost want to put scare quotes around "individual" here) and society. I would like to do that as well. It may seem strongly disanalagous at first, but close introspection of one's own mind shows that while there is something unified or unifying about conscious experience, the mind itself is made up of many different parts, each of which has its own motivations, goals, and behaviors. Each of us can be treated like a society in and of ourselves.

This is talked about in great detail in the meditation traditions that spread out from India, as well as in various psychotherapeutic modalities, from newer modalities like Internal Family Systems and Core Transformation to more traditional modalities like paychodynamics.

When you spend some time working with the parts through therapy and/r meditation, it starts to become clear that there us a far broader possibility space for the state of the whole bodymind system than one initially thought possible. Things can be both far better and far worse than one thought was possible. One can become aware of ways that your parts interact with each other that are quite surprising. For example--over the last week in my own meditation practice, I have discovered that there is a part of me which functions as a "bliss limiter"--it tamps down on a particular body experience of fizzy joy and pleasure which the Buddhist tradition calls "piti." It has been tamping this down this experience specifically because other parts were afraid of losing control and being unable to fulfill their functions and their goals. But I have found a way to allow the bliss limited to safely ease up a little bit without threatening too many other parts.

By analogy, I strongly suspect that there are aociety-wide modes of relating which are both far better and far worse than what we experience now. And just as clinging too hard to what works now in the individual status quo can lead the individual to severe depression, anxiety, suicidality, etc., I suspect that clinging to status quo methods of working with society can also lead us to dark places.

Singer's work is a little bit like the standard advice for dealing with negative mental health states. Giving to charity and changing to more prosocial personal habits and modes of consumption is rather like eating your vegetables, getting exercise, and trying to get enough sleep. They are indeed helpful and have been shown both scientifically and in many people's personal lives to improve things overall. However, they are limited in what they can do. There are other, very surprising ways of working which can have much more dramatically positive impacts.

Similarly, I suspect that there are things we can do as a society which can improve the overall state much more dramatically than we usually think possible. I think we can point to world history over the last 500 years for many such previous examples. (As well as many times where the local situation became much worse than people thought was possible quite suddenly! Which should be a warning!)

There's a useful concept in mathematics called a "local maximum."

One way to solve a problem is to simply do the thing that most directly relates to maximizing a desired attribute in ourselves and the world. This is roughly analogous to the "greedy algorithm" in computer science, and it's what Singer largely advocates for.

However, it usually results in getting stuck in local maxima--a space where any different incremental actions lead to a result that is worse than the result we're currently. It is often the case that in the total probability space of reachable solutions, there may be better solutions, but we will have to go through a temporary period of pain or reduced effectiveness to get there. But since we don't yet have that new solution--the problem isn't fully solved yet--we can't know for sure that we actually can get to a better solution than the local maximum that we've found.

There's a element of wisdom involved in apprehending that what we've found is merely a local maximum, not the global maximum. And there's a element of risk in striking out for the the higher peak across the valley. Call it a calculated risk or--if you want a drop of poetry--a leap of faith.

This criticism of Singer is essentially that he lacks the courage to step out from the local maximum that he's found. Whether it's a fair criticism or not depends on one's overall judgement of the situation.