A framework for constructing a cognitive model of users' information searching behaviour is described. The motivation for the framework is to create explanatory and predictive theories of information searching to improve the design of information retrieval (IR) systems. The framework proposes a taxonomy of components for process models of the information seeking task, information need types and knowledge sources necessary to support the task. The framework is developed into a preliminary version of a cognitive theory of information searching by the addition of strategies and correspondence rules which predict user behaviour in different task stages according to information need types, facilities provided by the IR system and knowledge held by the user. The theory is evaluated by using claims analysis based on empirical observations of users information retrieval and by a walkthrough of an IR session to investigate how well the theory can account for empirical evidence. Results show that the theory can indicate the expert strategies which should be followed in different task contexts but predictions of actual user behaviour are less accurate. The future possibilities for employing the theoretical model as a tutorial advisor for information retrieval and as an evaluation method for IR systems are reviewed. The role and potential of cognitive theories of user task-action in Information Retrieval and Human Computer Interaction are discussed.