Ability of solving mathematical non-routine problems among Malaysian students remains at a low level, therefore there is a need to introduce a new and effective teaching strategy for the teaching and learning of mathematical problem solving. This study aims to determine the effectiveness of Mayer's problem solving Model with Visual Representation (MMVR) teaching strategy in enhancing Year 4 students' mathematical problem solving ability. A sample of 175 Year 4 students were drawn using a convenient sampling technique with two classes assigned as experimental groups namely Mayer's problem solving Model (MM) group (n = 57) and MMVR group (n = 58), and the other one as the control group (n = 60). The MM group was given Mayer's problem solving Model (MM) teaching strategy treatment, while the MMVR group was given Mayer's problem solving Model with Visual Representation (MMVR) teaching strategy treatment, respectively. The control group on the other hand did not undergo any additional intervention from researcher. The results of one-way ANCOVA analysis reveals that there is a statistically significant difference in the mean of the posttest score in MPSAT between the three groups [F (2, 171) = 291.44, p < 0.0005], with the adjusted means of posttest scores of MPSAT for MM, MMVR and Control groups were 105.84 (SE = 0.94), 116.14 (SE = 0.93), and 85.94 (SE = 0.89) respectively. The results suggest that mathematics teachers can modify their current pedagogy following Mayer's problem solving Model with visual representation teaching strategy to increase students' performance in problem solving.