With increasing penetration of intermittent renewable energy sources into the electric grid, conventional thermal power plants are being forced to cycle their load and operate under low-load conditions much more frequently. A high-fidelity dynamic model of a natural gas combined cycle (NGCC) power plant with rigorous equipment level submodels is developed in this work. A model of the gas turbine (GT) for estimating its performance under off-design conditions, a thermo-hydraulic model for the heat recovery steam generator (HRSG), and a model of the steam turbine (ST) with moisture detection and model adaptation capability are also developed. Five novel configurations are proposed for controlling the main steam and reheat steam temperatures while avoiding "spraying to saturation" during fast load-following by considering different final high-pressure superheater (HP SH2)/reheater (RH) arrangements (in series or in parallel) and attemperation strategies (single-stage, two-stage, and damper-assisted attemperation). Load-following operation is studied under two operational strategies, i.e., the industry-standard coordinated control strategy and dynamic optimization. Dynamic optimization is used to maximize the plant efficiency while satisfying the operational constraints and state transition constraints. It was observed that while more than one configuration can avoid spraying to saturation and maintain the steam temperatures within their limits even during very fast load transients, their efficiency can greatly vary and that dynamic optimization, when feasible, can lead to superior efficiency than the coordinated control system. The parallel HP SH2/RH configuration using the damper-assisted attemperation strategy shows the highest efficiency.