Hah, yeah, that's the other thing. I'm not entirely sure what the inciting incident should actually be. I'd really like to tie it in to an act of terrorism (long story short: "cyborgs" created to supercede human workforce break programming, demand rights, splinter groups start blowing things up to make a point. Cyborg workers ran the nuclear plant and cause some kind of explosion/accident.)
Possibly penetration of the reactor core, causing meltdown?
This scenario is actually an awful lot easier to use to create an inciting incident than some kind of external terrorist attack.
At Chernobyl, essentially what happened (and I am simplifying here because I studied more about the after effects than the event itself), was the engineers wondered what would happen if you took the control rods out of the core. It was an experiment to see what would happen if the electricity failed - could the reactor still function on low power.
This caused the core to overheat, and then when they pressed the emergency shutdown, it caused the displacement of coolant and so the reactivity was concentrated in the lower part of the core. Then a series of explosions occurred.
The BBC have a really good time analysis of what happened here.
The explosions caused a radioactive 'plume' to be discharged (and now we're into the bit I know more about!) - this was affected by the wind, and therefore the question you're asking about exclusion zones is greatly affected by the prevailing weather. You need to look up the dominant wind directions and speeds to have any idea of how far out harmful levels of radiation will reach. Don't forget, the incident at Chernobyl was initially covered up and denied - it was only because Scandinavian planes (I want to say Swedish but that might be wrong) started to detect high levels of radiation that the Government in Moscow came clean - that was several (three I think) days after the incident. There are parts of Scandinavia which are still affected by it today - farms with strict controls on what they can produce for example.
In terms of the exclusion zone many years later, the interesting thing about Chernobyl is that the dangerously high levels of radiation only occur where the plume hit the ground - so it isn't a clear cut thing. It isn't like a neat isoline diagram which shows neat rings where it is and isn't dangerous. Effectively, there are large swathes of land where the impact was nil, and equally large swathes where it is lethal to be in there. The trouble is, you can't see where it changes. The exclusion zone around Chernobyl is basically a large nature reserve, and the exclusion of humans has far outweighed the negative impact of the radiation on wildlife (though clearly I'm not saying it was a good thing!).
There is some speculation that the lack of negative impact on wildlife is down to them being much shorter lived than humans, but there is also some evidence suggesting the wildlife may have some mechanism for detecting where the highest levels of radiation are and avoiding those areas. Someone on my MSc course did their thesis on the latter hypothesis, though of course an MSc thesis doesn't take place over a long enough time scale to produce anything like the evidence you'd need to prove something so complex with so many variables.
So the short answer is, it would be perfectly feasible to have a relatively small exclusion zone in some areas, because the plume didn't touch the ground there, as long as you also made it clear that there was no contamination of the groundwater system (assuming you want to do things like have people grow crops, or use aquifers for their drinking water supply). Equally, the effects could be huge a long way out due to the wind carrying harmful radiation in the upper atmosphere. Which all means, whatever fits your story is fine I think, as long as you justify it. You could have cyborgs go in and test for the radiation boundaries for example, to work out where the plume hit the ground and where it's safe to go?
If you run a search on Google Scholar using something like 'wildlife Chernobyl exclusion zone' you'll get a whole load of papers and abstracts which talk about this stuff, and which for the most part are peer-reviewed so relatively trustworthy.