Need help with FIMIX-PLS method, problematic results?

[SmartPLS user]

Can someone tell me whether the following segmentation results are correct or not?
Model description: There were three exogenous latent variables (say E1, E2, E3) and two endogenous latent variables ( one is SAT and the other one is LOYALTY)

When I run the FIMIX segmentation using SmartPLS 3.0, (number of segmentation is 2), I got the following FIMIX R Square values for the final target variable LOYALTY in each segment:

Segment 1: R square value for LOYALTY is 0.495; size: 0.620

Segment 2: R square value for LOYALTY is 1.000; size: 0.380, the standardized path coefficient from SAT to LOYALTY is also 1.000. in segment 2.

My question: If there is a strong reason to believe there are two heterogeneous segments:
1) Is this R square value (1.000) for LOYALTY in segment 2 problematic or credible? The R square value of 1.000 seems suggest that the three exogenous latent variables E1, E2 and E3 completely explained the variance of LOYALTY?
Note: The entropy statistic (EN) is also 1.000, For me, the result is just too dramatic to be true....

2) The standardized path coefficient of 1.000 between SAT and LOYALTY seems suggesting that and the SAT completely mediated the impacts of E1, E2, E3 on Loyalty, is this result problematic or credible?
Any comments or advice will be highly appreciated.