Evolving Enactivism covers a lot of ground, from basic perception to autobiographical memory. The basic thesis is that all these cognitive activities are "contentless", or free of representations (which are defined as mental models of the environment that have veridicality conditions). Hutto's approach is to summarize claims made by cognitive scientists and philosophers on relevant issues and criticize these claims. This approach is useful, in a number of ways. Readers can access a variety of references, discover what topics within embodied cognition interests them most, and easily branch off into these directions through Hutto's references. Also, readers can get a better sense of the fields overall and a more nuanced understanding of each argument, since they are in meaningful juxtaposition to each other.
However, there are problems to Hutto's approach as well. Because he covers so many topics within this one book, his criticisms of opposing arguments are sometimes hurried and unconvincing. In the end, he has very few points that challenge traditional representational views on cognition. He repeats these points over and over again throughout the chapters, in different phrasings. He doesn't examine the possible counterarguments to his own arguments and supports them often using general claims (he rarely uses detailed empirical evidence).
Furthermore, he doesn't even go into theories in the extended mind literature to support his claims, until the epilogue (and there, he reviews these theories only briefly). I believe theories in the extended mind, including Gibson's ecological affordances and Noë's sensorimotor contingency theory, offer great conceptual resources that Hutto could've drawn upon to support his thesis. In the end, by merely sticking to his few points when responding to the various challenges to radical embodied cognition, he make it seem that this non-traditional view differs merely semantically from the traditional representationalist view. On my reading, it seems Hutto's version of embodied cognition is similar to representationalism, and even identical to it when concerning "offline" cognitive activities (e.g. imagination and memory). The only difference Hutto maintains on these latter issues is that his version of embodied cognition eschews the veridicality conditions that representationalist views maintain.
If readers are interested in Hutto's book and the fields of embodied cognition and enactivism, I would instead point them to Chemero's "Radical Embodied Cognitive Science" (2009) and Rowlands' "The New Science of the Mind: from Extended Mind to Embodied Phenomenology" (2010). These two books present positive theories for embodied cognition that substantively differ from those in traditional representationalism. These two books do not shy away from philosophical detail and empirical studies. They give convincing accounts on how embodied cognition can explain information processing (and thereby perception and other cognitive activities) in a non-representationalist way, which is something I think Hutto fails to do.
The one point I appreciate that Hutto makes, and which other books on embodied cognition that I've read do not, is that non-representationalism in cognition entails that cognitive activities (e.g. perception, imagination, and memory) do not have an objective teleology. Traditional representationalist accounts assume that these cognitive activities have the inherent function to be accurate to "external reality." In my metaphysical view, there is no determinate, objective reality, so this lack of objective teleology sounds true to me. It is pleasing existentially that we have freedom to choose our own ends, and our cognitive abilities are there to aid in pursuing them.