REA modeling technique is very important in generating a well-structured database which will be able to affect the performance of any accounting information system. The author has implemented quantitative analysis by generating mathematical algorithms to compare efficiency between conventional and Agent Centric REA Modeling. However, most of the end users do not really understand the representation of mathematical algorithms. Hence, agent derivation mathematical algorithms have been converted into a simulation tool which plays a vital role in performing qualitative analysis, which involves the analysis towards agent centric simulation system's usability from users' perspective. The amount of 20 third year students from department of Science and System Management, Fakulti Teknologi dan Sains Maklumat (FTSM), Universiti Kebangsaan Malaysia were chosen randomly as qualitative comparison's sample, based upon their knowledges in REA modeling. Respondents will be asked to test both system prototypes translated from event approach and agent approach. Results towards mathematical formulas' simulation test shows that the total amount of redundant entity nodes generated by using conventional REA modeling technique always higher than the suggested methodology which proposes agent approach. Besides that, in the aspects of qualitative comparison, results clearly shows that 100% of the sample agreed that the prototype system was which translated from agent approach is able to generate final output faster and more accurate, besides requiring the minimum modeling steps. In addition, agent approach translated system prototype provides the maximum and optimum level of users' comfort; satisfaction and ability to increase REA modeling know ledges for the system's users.