Initially, new nanocomposite reverse osmosis (RO) membranes containing polypyrrole (PPy) coated on multiwalled carbon nanotubes (MWCNTs) were synthesized. Then, the synthesized membranes were prepared to model flux measurements using artificial neural networks (ANNs). Raw and oxidized MWCNTs were coated with the polypyrrole and then various amounts of them were mixed in m-phenylene diamine solution to prepare polyamide RO membranes using the interfacial polymerization method. The contact angle, surface morphology, surface roughness, salt rejection, water flux and fouling efficiency were investigated. Based on the results, water flux and fouling performance were enhanced. For both the raw and oxidized MWCNTs-PPy blended membranes, water flux increased from 21.5 to 30.4 and 34.3 L/m(2) h, respectively. Also, antifouling properties were improved by embedding the polypyrrole nanocomposites particularly in 0.002 wt% oxidized MWCNTs-PPy membrane. Using ANNs, the water flux measurements were modeled with input parameters including temperature (temp degrees C), trans-membrane pressure (TMP), time (h) and MWCNTs-PPy concentration (raw and oxidized). According to the modeling data, Mean square error was minimized and correlation coefficient was higher than 97%.