This research was a quasi-pretest-posttest design experimental study with experimental and control classes. The experimental class applied the VAK model and the control class applied the guided inquiry model. The research was conducted on fifth-grade students of primary schools in Surakarta in the academic year 2018/2019 with a sample of 114 students selected by random sampling technique. The research data was obtained by a narrative writing test instrument that had been validated by linguists and evaluation experts before tested. The results showed the mean score of narrative writing skills gained in the experimental class was 0.44 compared to the control class of -0.30. The results of the independent sample t-test showed a significant difference between the experimental class and the control class with a t-value of 25.155 and a sig. (2- tailed) value of 0.000<0.05. The effect-size calculation showed that the VAK model had a high effectiveness in learning narrative writing skills, with a value of d=7.58. Based on the theory, related to learning style (visualization, auditory, kinesthetic) could enhance the increase in linguistic intelligence as an internal factor affecting narrative writing skills. Based on research findings and theories, it can be synthesized that in terms of linguistic intelligence, the VAK model is effective in improving narrative writing skills.