This paper introduces a practical, interactive simulator of the limit order book for large-tick assets, designed to produce realistic execution costs and P&L for strategy evaluation. The authors extend the queue-reactive (QR) model with three key modifications: a state projection onto spread and volume imbalance for robust estimation, a non-exponential inter-event time distribution that captures the clustering of events at exchange round-trip latency (revealing latency races among participants), and a market impact feedback mechanism that accumulates signed trade flow through a power-law decay kernel, reproducing concave impact during execution and partial post-trade reversion. The simulator is validated on four S&P 500 stocks using two years of data, showing that it reproduces marginal statistics such as event-type frequencies, imbalance before trades, and volatility, while addressing the limitations of the original QR model in inter-event time distribution and market impact. Two case studies—a mid-frequency signal-based strategy and a high-frequency imbalance strategy—demonstrate that the simulator yields realistic behavior where profitability becomes highly sensitive to execution parameters, making it a valuable tool for practitioners. The approach is presented as a practical recipe: project the book onto a tractable state, estimate conditional distributions from data, validate against empirical statistics, and adapt to the asset or use case.