deep learning LSTM time series MISO

Dear Sir

Looking at the deep learning toolbox I see an example for time serie forecast named (Time Series Forecasting Using Deep Learning) using LSTM. In this example one time series is used and forecasted x steps ahead.

I am wondering if and how it would be possible to use the deep learning, LSTM, to have one time series to predict/forecast output and a second time series use as input. Similar like with NARX (MISO like). And be able to forecast the next outputs (x-steps ahead) by using new inputs.

Many thanks