Understanding how temporal sequences are learned and processed is of fundamental importance in understanding cognitive processes. The current proposal presents a model of sequence learning and processing which seeks to explain how these phenomena might work in the context of biologically-justified learning mechanisms and broad topographical connectivity patterns. The model works on the hypothesis that the primary role of excitatory feedforward connectivity is the hierarchical recruitment of sequence representations. The hypothesized role of excitatory lateral connectivity is primarily to form auto-associative links between representations at the same level of abstraction, while the role of excitatory feedback connectivity is to propagate predictions back down the hierarchy, thereby disambiguating noisy and incomplete input from below. A preliminary version of the model demonstrating the role of feedforward connectivity in hierarchical recruitment learning is presented on a temporal XOR-style task. The problem is treated as one of sequential feature binding. The model is novel in its ability to learn sequences in one shot in an unsupervised manner, using simulated spiking neurons and biologically- plausible learning mechanisms. The model also makes novel predictions regarding the physiology of cortico-cortical connectivity and its psychophysical ramifications. Extensions of the model for future simulation and research are then discussed.