Nonlinear Model Predictive Guidance (NMPG) based on optimal control (trajectory optimization) is an extremely promising method for achieving optimal and precise guidance. Event-triggered NMPG offers a dynamic cycle strategy where updates to the optimal guidance law are made only when the deviation between the actual state and the optimal reference trajectory exceeds a set threshold, thus conserving computational resources effectively. To further minimize event-triggering occurrences, we propose embedding an optimal parameter estimation step into it. With increasing sampled flight-state data over time, the motion model for NMPG progressively converges towards the actual one, thereby yielding more precise open-loop guidance laws. The simulation results of an ascent guidance problem show that the proposed method has higher guidance accuracy, a shorter simulation time, and fewer guidance law update times than traditional iterative guidance (IG) and general NMPGs.