Text compression models are extremely important and considerably useful when they are irreversible and noiseless. A number of these models were repeatedly examined and tested until convincingly refined to keep only most suitable ones of them. Those selected models can reversibly compress English text files, programs, and Arabic text files with optimum results and processing- measures in terms of speed, size and accuracy . The proposed research also presents the criteria and the methodology of designing and implementing the interactive and user - oriented computer system which functions on microcomputers. Considerable part of the research was dedicated towards finding a conclusive single model to cover all types of text files, programs, and Arabic text. However, a conclusive single system was achieved based on collective models, with a quite successful Arabic reversible compression capabilities. The proposed system was implemented with the Turbo-Pascal 5.5 version on the IBM - compatible Microcomputer.