His work also features lots of surveillance themes (again, likely due to the nature of his psychosis) and I can't think of anything more relevant to modern society. We've now reached the stage of technology where it's feasible for us all to be living in a panopticon.
Another major theme in PKD's work is "what makes us human" and we're getting close to producing LLMs that seem to behave as though they are human. If we give an AI a sense of personal identity with some fake memories, then would we be able to tell that they weren't human and what would they "think" about it?