We previously reported relationships between predicted supplies of MP, MP-Met, and MP-Lys (NRC, 2001) and yields of milk and milk protein. The original database consisted of model evaluations of 464 diet treatments from experiments published in the Journal of Dairy Science (JDS) from 1976 through 2003. Results indicated yields of milk and milk protein were more accurately predicted by supplies of the first limiting AA rather than by supplies of MP. To establish a more robust database, an additional 550 diet treatments from experiments published in JDS in 2004–2014 were evaluated. For the current analysis, the following criteria were imposed: NE_L allowable milk ≥ MP allowable milk, MP allowable milk within ± 6 kg of actual milk yield, MP balance between −250 to 100 g/d, Lys ≤7.2% of MP, and Met ≤2.4 of MP. To develop the regression equations for MP-Met, and to help ensure that Met was more limiting than Lys, the Lys/Met ratio in MP had to be ≥3.20/1. To generate the regression equations for MP-Lys, the Lys/Met ratio in MP had to be ≤3.15/1. It was preferred this ratio be ≤3.00/1, but that was too restrictive for the available data set. The PROC REG procedure of SAS was used to generate the regression equations. The resulting regression equations describe the relationship between measured milk yields and MP, MP-Met and MP-Lys supplies: MP (n = 455): y = 0.01397x + 1.76875, R² = 0.74; MP-Met (n = 260): y = 0.84054x − 1.18403, R² = 0.78; and MP-Lys (n = 29): y = 0.25722x − 1.48198, R² = 0.70. The equations describing the relationship between milk protein yields and MP, MP-Met and MP-Lys were MP (n = 455): y = 0.47131x − 67.02527, R² = 0.82; MP-Met (n = 260): y = 28.47206x − 178.08057, R² = 0.86; and MP-Lys (n = 29): y = 9.07155x − 217.19663, R² = 0.85. In contrast to the previous equations, these are linear in nature and biologically more correct. We conclude these updated Met and Lys regression equations more accurately predict relationships between MP-Met and Lys supplies and milk and milk protein yields, and can be used by NRC users to determine if milk and milk protein yields are potentially being limited by these 2 AA.