Using maximum likelihood, any of these can be compared with either the likelihood ratio test or AIC.  Using REML, the fixed effects must be the same; so `m5` is not comparable to the others.  However, interpretation is usually difficult when both fixed effects and random effects are changing, so in practice, most recommend changing only one or the other at a time.   

Pinhiero/Bates is the classic reference; it describe the `nlme` package, but the theory is the same.  Well, mostly the same; Doug Bates has changed his recommendations on inference since writing that book and the new recommendations are reflected in the `lme4` package. But that's more than I want to get into here.  A more readable reference is Weiss (2005), Modeling Longitudinal Data.