Predictions for TFTR current scaling experiments are investigated for four recent theoretically based models, to find support for empirical current scaling used for planning future fusion devices. The profile-consistent drift wave model is found to predict current scaling. None of three multimode models, including the Bateman Version 5.10 model, previously claimed to show current scaling for TFTR, succeeds in simulating measured temperature and density profiles for either high or low current experiments. Standard edge resistive ballooning mode models do not lead to Ip scaling of confinement, underestimating losses at low current by factors of 2 to 4