Nitrous oxide (N2O) measured simultaneously with di-nitrogen (N2) emissions from soils are greatly uncertain due to large temporal and spatial variations. This study aims to report N2O, N2, N2/N2O, 15N-N2O, wheat-maize annual grain yields, and yield-scaled N2O and N2O plus N2 emissions on the responses to different nitrogen (N) fertilizer rates in a winter wheat-summer maize cropping system. Furthermore, this study also seeks to determine controlling factors for N2O, N2, and N2/N2O emissions and significantly investigate the relationship between the soil-climate measured factors and 15N-N2O. Three N inputs and control treatments, 0, CK; 200, LN; 400, MN; and 600, HN kg N ha−1 year−1 were set since 1998. Direct measurement method has been used to quantify N2O and N2 emissions. Our results indicated that the effects of long-term N fertilization significantly increased N2O and N2 and also reduced N2/N2O emission ratios as described by exponential functions. Using structural equation modeling (SEM), NH4+, WFPS, NO3-, and DOC were revealed to be main controlling factors for N2O, while N2 by DOC, NO3-, WFPS, and temperature finally N2/N2O was positively related to temperature. Furthermore, the 15N-N2O was positively related to N2/N2O ratios, indicating that denitrification is the dominant process at the study site. The yield-scaled N2O emissions followed the order HM>MN>LN>CK, and they were 1.56, 1.47, and 1.07 times greater than CK, respectively. Total yield-scaled N2O plus N2 were in the order of CK>HN>MN>LN. N fertilization has shown strong impact not only on N2O, N2, and N2/N2O emissions but also on yield-scaled N2O and N2O plus N2 emissions. High agronomic nitrogen use efficiency (NUE), low yield-scaled N2O emissions, and low cumulative N2O plus N2 emissions were observed at 200-LN treatment, suggesting this rate to be an optimum and sustainable agricultural management practice with no significant crop yield reduction as compared to the current farmers’ practice of 400 kg N ha−1 year−1.