@Antonio You don't need to look for RMSEA as in the results of Smart-PLS, you SRMR which is an alternative to RMSEA and the recommended threshold for both of them is between 0 and .08. For reference, you can refer to the attachment.

Marco Antonio Cueto Daza, PLS-SEM (variance based) uses a different algorithm than ML-SEM (covariance based). Thus, PLS-SEM will not provide you with RMSEA or any equivalents. It has instead its own model quality measures such as GoF although this measure is also debated.