Humans will evacuate an area on the basis of enough radiological threat that one in a hundred gets an additional tumor sometime in middle age, or perhaps if one in ten neonates in the same area are born with some visible deformation.
Wild animals need a threat thousands or perhaps tens of thousands of times higher to actually threaten their ability to sustain themselves. All they need to keep the population up is replacement-level fertility on average, and most animals are physically capable of many, many times that level.
In this higher dose scenario lifespan is reduced so much that it starts to impinge on their reproductive capabilities, which are usually food/habitat limited by human influence (or previously to humans, by predators and disease as well).
Humans made a tradeoff towards k-selection: in order to support these huge skulls & brains, we made childbearing an epic process that takes a decade, and made childbirth quite dangerous. For something like heavily r-selected fish, two parents might produce ten thousand eggs, and only two of them need to survive to breed to sustain the population. Remove all the other things that kill this example fish, and each generation is five thousand times the size of the previous generation. Most mammals are somewhere between fish and humans.
Free up the habitat and let them reproduce ad libitum, and over all but the worst-hit areas, wild animal populations will grow rapidly because the radiation is less of a population stressor than the people, their cars, their pavement, their agriculture, and their fishing nets were.
This was not an expected result. The efficacy of Chernobyl as a wildlife preserve was kind of a wake-up call for ecologists about how deep into the anthropocene epoch we actually are.