Abstract: The trapping of positrons at the surface of a material can be exploited to study quite selectively the surface properties of the latter by means of positron annihilation spectroscopy techniques. To support these, it is desirable to be able to theoretically predict the existence of such positronic surface states and to describe their annihilation characteristics with core or valence surface electrons in a reliable way. Here, we build on the well-developed first-principles techniques for the study of positrons in bulk solids as well as on previous models for surfaces, and investigate two schemes that can improve the theoretical description of the interaction of positrons with surfaces. One is based on supplementing the local-density correlation potential with the corrugated image potential at the surface, and the other is based on the weighted-density approximation to correlation. We discuss our results for topological insulators, graphene layers, and quantum dots, with emphasis on the information that can be directly related to experiment. We also discuss some open theoretical problems that should be addressed by future research.