A strategic model of Nucleopolyhedrovirus (NPV) genetics is developed and applied to co–occlusion and the dynamics of the few polyhedra (FP) phenotype. Co–occlusion is the incorporation of wild–type and mutant viruses in the same occlusion body as a strategy to deliver genetically modified viruses as an insecticide in a way that minimizes their persistence in the environment. The model predicts that the persistence of the modified virus depends most critically on the number of viruses that infect an occlusion body–producing nucleus, the number of occlusion bodies typically ingested by an insect, and the number of viruses that cross the gut wall. FP viruses occur at high frequency in cell culture where they have a replication advantage over wild–type virus. They cannot, however, produce viable occlusion bodies, but it is shown that under certain circumstances they can persist in natural populations as parasites of wild–type viruses. FP viruses have not as yet been identified in natural NPV populations, and possible reasons for this are discussed. We suggest that the models developed here can also help in studying potential recombination between engineered and wild–type viruses in the environment.