Salaries are in essence what we are willing to pay one another for each other's time and effort. In your example, people are willing to pay a person 50 times as much for drafting a contract than for taking care of a disabled person. I don't see anything inherently wrong with this. There could be various reasons explaining this disparity, ranging from a lack of availability of people who can do that, to that job potentially being more difficult in some ways, and probably also down to it being more important in some ways, e.g. directly affecting the lives of thousands of people. Whatever the reasons are, the people who choose to pay the fee were probably unable to negotiate a lower fee for a similar level of work.
On the other hand, the question of "what we should value" is much more amorphous and I'd be interested to hear of ideas on how to model and practically implement that, if you have any. There only alternative frameworks I'm familiar with are communism, with its well known share of problems and "time banks" which I've seen working decently well at a local level, but don't tend to scale.
Under the view of the grandparent, the metric/proxy of salary has become so utterly meaningless because of modern day business behavior that it is no longer a meaningful thing to pursue (or never was).
I personally also have some problems with your view. According to you a corporated lawyer earning $500/hour by drafting contracts is 50x as valuable to society as someone earning $10/hour caring for disabled people, which I think is very untrue. What "the market" values and what we as people should value only has a partial correlation.
Another example, a lot of people value their family and children above all else, while having children leads to almost zero reward from "the market".