Ozone is one of the most significant secondary pollutants with numerous negative effects on humanhealth and environment including plants and vegetation. Therefore, more effort is made recently bygovernments and associations to predict ozone concentrations which could help in establishing betterplans and regulation for environment protection. In this study, we use two Artificial Neural Networkbased approaches (MPL and RBF) to develop, for the first time, accurate ozone prediction models, onefor urban and another one for rural area in the eastern part of Croatia. The evaluation of actual againstthe predicted ozone concentrations revealed that MLP and RBF models are very competitive for thetraining and testing data in the case of Kopaˇcki Rit area whereas in the case of Osijek city, MLP showsbetter evaluation results with 9% improvement in the correlation coefficient. Furthermore, subsequentfeature selection process has improved the prediction power of RBF network.