Biological and physical interactions in unsaturated soil, the vadose zone, have received a surge of research interest over the past several years. This article reviews recent research, focusing on the limitations imposed by the complexity of soil, the use of model systems to understand processes, new technologies, and the understanding of how biology changes soil structure. Research using model systems to mimic natural structure, such as rough planar surfaces or packed columns, has made it possible to demonstrate and quantify microbial interactions at very small spatial scales, including the coexistence of competing microbes and the invasion of soil pores by organisms that should be too large to fit. It is now possible to see inside soil at micrometer resolution in three dimensions, either by the use of noninvasive imaging techniques on intact soils or a model transparent soil with the same refractive index as water. Soil biology also changes soil structure. Techniques from engineering such as fracture mechanics and rheology have measured enhanced particle bonding, dispersion, and aggregation caused by root and microbial derived exudates. Models of soil structure dynamics are beginning to use these data. Concurrent research on naturally structured soil is essential, but using model systems that allow for the application of material science approaches or the detection and modeling of specific processes will enable the building of complexity by piecing together simpler systems. A major challenge for future research is gaining a quantitative understanding of how soil biology changes structure and incorporating this knowledge with studies of soil biodiversity, microbial functions, and root–soil interactions. Upscaling from microbial processes at micrometer resolution to the whole plant, field or catchment presents an even greater challenge.