The "O'Nan-Scott approach" to describing structure for finite primitive permutation groups is now a standard way to understand and analyse these groups. I will discuss an exploration made with Peter Neumann and Simon Smith of the extent to which a similar approach might be successful in the realm of infinite groups.