This article will discuss learning to write short stories with the picture and picture model. Is there any significant difference between the ability to write short stories using the picture and picture model and that which does not use the model? The study was conducted using quasi-experimental design with the pre test-post test of control group design. The study was conducted in class XI of SMK 2 Cilaku. Data collection techniques use tests and questionnaires. Reliability testing is done with the Cronbach alpha coefficient formula. The calculation results show the reliability value of 0.873 which is greater than the coefficient value of 0.6. Data analysis techniques were carried out using the post-test and t-test results scores. The results showed a significant difference in the short story post-test scores of the control group and the experimental group. In the control group there were 3 (three) students who scored low, 8 (eight) students scored medium, and 3 (three ) students got a high score. In the experimental group 1 (one) student got a low score, 11 (eleven) students scored medium, and 7 (seven) students got a high score. This shows that there are significant differences in short story writing skills between the experimental group and the control group. The results of the t-test of the pre test score and the post test of the experimental group had a significant 2-tailed 0.009 smaller than 0.05, which showed that Ho was rejected with the difference the experimental class average value of 72.9474 higher than the control class of 70.2500, based on these results that the picture and picture learning model can improve the ability to write short stories at SMK Negeri 2 Cilaku.Keywords: implementation, picture and picture model, writing short story