The article discusses three areas that appear neglected or underdeveloped in current treatments of scalar timing theory (SET). In particular, questions about where variance in the SET system comes from, and how memory and decision processes operate within SET are discussed. The article suggests a number of possible experiments with humans, some based on pilot work which is described, that may clarify all three areas to some degree. Methods derived from conventional studies of memory are suggested as providing techniques for investigating the operation of memory and decision processes within the SET model, both areas previously considered somewhat inaccessible. In general, the tripartite division of SET into clock, memory, and decision processes is advocated as a useful general framework for studying timing, including questions related to its neurobiological basis, whether or not data always conform to SET predictions, although more needs to be known about how all three parts of the SET system operate.