OK. That helps. 1) correct. So long as you're far from the tipping point herd immunity %, no, 1 more unnvaccinated person doesn't increase risks much for other unvaccinated people and the vaccinated people should mostly be fine anyway.
Part 2 is conditional. The risk of infection spreading is very non-linear. There's a tipping point, that's different from place to place (population density is a factor) and from disease to disease. Let's say the magic number for a certain population is 78% Then (and I'm making up these numbers) if 90% get the vaccine, 0.1% get sick in a given year.
85% -> .2% sick.
80%-> .5% sick.
78%->1% sick.
77%->5% sick
76%->10% sick(epidemic among the unnvaccinated)
Again, I made up the numbers, but that's what I mean by 'tipping point'. If enough member of the herd are immune, we have 'herd immunity', meaning the disease is unlikely to spread through the herd because a sick person isn't likely to run into another unnvacinated person. When the number of anti-vaxxers gets too high, we lose herd immunity. But that assume a random distribution of vaccinated and unvaccinated people.
This effect is compounded by the fact that anti-vaxxers tend to be in the same social groups. So if you have a compromised immune system and you pass someone on the street that's not vaccinated but is in a group that has herd immunity and unvaccinated people are evenly distributed, there's not much of a chance they will be infected. But if a group of unvaccinated people all have their kids in the same school, there's a much higher chance when you walk by them that they will be infected.
Proof is in the news with these outbreaks.