Two independent chemistry-transport models with troposphere-stratosphere coupling are used to quantify the different components of the radiative forcing (RF) from aircraft emissions of NOx, i.e., the University of L'Aquila climate-chemistry model (ULAQ-CCM) and the University of Oslo chemistry-transport model (Oslo-CTM3). The tropospheric NOx enhancement due to aircraft emissions produces a short-term O3 increase with a positive RF (+17.3 mW/m2) (as an average value of the two models). This is partly compensated by the CH4 decrease due to the OH enhancement (−9.4 mW/m2). The latter is a long-term response calculated using a surface CH4 flux boundary condition (FBC), with at least 50 years needed for the atmospheric CH4 to reach steady state. The radiative balance is also affected by the decreasing amount of CO2 produced at the end of the CH4 oxidation chain: an average CO2 accumulation change of −2.2 ppbv/yr is calculated on a 50 year time horizon (−1.6 mW/m2). The aviation perturbed amount of CH4 induces a long-term response of tropospheric O3 mostly due to less HO2 and CH3O2 being available for O3 production, compared with the reference case where a constant CH4 surface mixing ratio boundary condition is used (MBC) (−3.9 mW/m2). The CH4 decrease induces a long-term response of stratospheric H2O (−1.4 mW/m2). The latter finally perturbs HOx and NOx in the stratosphere, with a more efficient NOx cycle for mid-stratospheric O3 depletion and a decreased O3 production from HO2+NO in the lower stratosphere. This produces a long-term stratospheric O3 loss, with a negative RF (−1.2 mW/m2), compared with the CH4 MBC case. Other contributions to the net NOx RF are those due to NO2 absorption of UV-A and aerosol perturbations (the latter calculated only in the ULAQ-CCM). These comprise: increasing sulfate due to more efficient oxidation of SO2, increasing inorganic and organic nitrates and the net aerosols indirect effect on warm clouds. According to these model calculations, aviation NOx emissions for 2006 produced globally a net cooling effect of −5.7 mW/m2 (−6.2 and −5.1 mW/m2, from ULAQ and Oslo models, respectively). When the effects of aviation sulfur emissions are taken into account in the atmospheric NOx balance (via heterogeneous chemistry), the model-average net cooling effects of aviation NOx increases to −6.2 mW/m2. Our study applies to a sustained and constant aviation NOx emission and for the given background NOy conditions. The perturbation picture, however, may look different if an increasing trend in aviation NOx emissions would be allowed.