There were a lot of things that went wrong in 3MI. Many of the lessons learned from that were incorporated into future designs. But one thing that went very right was that there was defense in depth, so that a N different things would have to go wrong to create a nuclear disaster. And in this case the number of failures was less than N. That's an engineering and regulatory success story.
A large amount of radioactive krypton gas was "vented", meaning it was released to spill down to the river and gas anyone who lived nearby. There was no tracking, so we don't know who or how many were exposed, or how much.
But this is a night-and-day comparison with, say, Chernobyl where the core was exposed and burning unmitigated for nine days. Many more orders of magnitude more release of radiation.
The nuke navy is always cited as having no incidents, but that doesn't pass the smell test. Military failures are easily classified and buried.
If Chernobyl had secondary containment, the burning fuel rods would not have been exposed directly to atmosphere. Basically, if you have a fire emitting toxic soot it's a lot better to have this fire happen in a concrete dome versus totally exposed.
We got super super lucky. And there's some debate about how bad the accident was with regards to NRCs monitoring.
Frankly, the whole plant was a disaster in the making. There was tons of warning lights and other systems but they were essentially useless because they constantly flashed and for poorly understood reasons.
3mile island is an excellent engineering study of what not to do with monitoring. We got very VERY lucky it was as small as it was.