The annual NO2 concentrations in many European cities exceed the established air quality standard. This situation is mainly caused by Diesel cars whose NOx emissions are higher on the road than during type approval in the laboratory. Moreover, the fraction of NO2 in the NOx emissions appears to have increased in modern diesel cars as compared to previous models. In this paper, we assess 1) to which level the distance-specific NOx emissions of Diesel cars should be reduced under real driving conditions to meet established air quality standards and 2) if it would be useful to introduce a complementary NO2 emissions limit. To this end, we develop a NO2 pollution model that accounts in an analysis of 9 emission scenarios for changes in both, the urban background NO2 concentrations and the local NO2 emissions at street level. We apply this model in a case study to the city of Antwerp, Belgium. NOx emission reductions reduce the regional and urban NO2 background concentration while a high NO2 fraction increases the NO2 concentration only close in big cities. In a busy access road to the city centre, the average NO2 concentration can be reduced by 23% if Diesel cars emitted 0.35 g NOx/km instead of the current 0.62 g NOx/km (2010). Reductions of 45% are possible if the NOX emissions of Diesel cars decreased to the level of gasoline cars (0.03 g NOx/km). Our findings suggest that the Real-Driving Emissions (RDE) test procedure can solve the problem of NO2 exceedances in cities if it reduced the on-road NOx emissions of diesel cars to the permissible limit of 0.08 g/km. The implementation of a complementary NO2 emissions limit may then become superfluous. If Diesel cars continue to exceed by several factors their NOx emissions limit on the road, a shift of the vehicle fleet to gasoline cars may be necessary to solve persisting air quality problems. ; JRC.H.2 - Air and Climate