That said, there are a lot of ways to throw money at a problem and produce very little in the way of results. The education establishment is very, very good at it.(That would be a rant for another day - see https://news.ycombinator.com/item?id=34634210 for a flavor.)
And about crime, the issue of targeting by police is not necessarily the neighborhood, but the person. When police see someone "who doesn't look like they belong", that person gets targeted. As a result black people in affluent neighborhoods get stopped a lot more than white people in the same neighborhoods. So even if black and white kids are doing bad things in similar amounts, the blacks are going to get arrested for it at a far higher rate.
This is not just a random conspiracy theory. See https://www.ojp.gov/ncjrs/virtual-library/abstracts/study-ra... for an example of research on racial disparities of how policing is done in Los Angeles. Many other cities have similar problems.
That the problem is not just limited to one bad police department is shown by a simple statistics. In surveys, blacks and whites do illegal drugs at similar rates. But blacks are arrested and charged for that at several times the rate that whites are. And so the arrest and jail statistics make it look like blacks are doing far more drugs than whites. But our best evidence is arrest records are a very severely racially biased sample of what is actually happening.
That said, I'm mostly in agreement with you on affirmative action. But there is one data point that shows a flaw in your argument. It is easy to argue that policies that helped a half-Kenyan kid get into Columbia for undergraduate and Harvard for law school won't help American blacks very much. But Barack Obama went on to become the first black US President. And the symbolism of that seems to be very important for inspiring US blacks in general. So even though it doesn't seem to me like it should matter, in practice it seems to have.