That depends on what exactly you mean by "Trojan sub moons"
If you mean one of the co-orbital bodies has an additional smaller body trapped in each of its leading and trailing Trojan points with respect to the primary, then no. The other co-orbital body will make close passes by each of them twice during each orbit-swap cycle, making those points unstable.
If, however, you mean that you want one of the co-orbital bodies to itself have moons orbiting it, one of which is in a Trojan orbit with respect to the other, then it may be possible--but only on a very large scale. I.e., this might work with co-orbiting planets and a pair of very small, close-in moons--or planet-sized co-orbiting moons of a gas giant and a pair of still-very-small, close-in sub-moons. The co-orbiting bodies would need to have their orbits tuned to maximize their separation at closest approach during a swap, and the sub-moons would need to be well within to the Hill sphere of their primary during said closest approach.