This paper examines six alternative planning/implementation (P/I) models identified from a review of the literature and then discusses the empirical P/I models observed in operation at three field sites where groups are working to create new, nontraditional schools. Separate sections of the paper define and clarify the functions of P/I models, suggest criteria for judging good planning and implementation, discuss five underlying dimensions of P/I models, apply these dimensions to six "classic" P/I models, discuss the empirical P/I models observed at three field sites, and make some general predictions about the types of P/I models that are likely to be effective under particular conditions. Much of the paper is devoted to discussion of six "classic" P/I models; these include the"muddling through" model, the "natural development" model, the engineering/architecture model, the rationalized planning model, the simulation model, and the adaptive planning model. (JG)