Eddying oceanic regions adjacent to continental shelves typically exhibit complex flow structure and, while the broad-scale ocean surface is well observed by satellites, the ocean's fine-scale flows and complex vertical subsurface are typically undersampled. The combination of oceanographic observations and numerical models, through data assimilation, produces data-constrained, four-dimensional fields that support a range of ecological and economic needs. Data assimilation makes use of a numerical model to place observations in dynamical context, yet it is particularly challenging for highly variable oceanic flows with sharp spatial gradients and where observations are sparse in space and/or time relative to the ocean's scales of variability. Here we use observing system experiments to assess the impact of recently-available observations from the Surface Water Ocean Topography satellite (SWOT, observing the ocean's fine-scale height structure) and from the Fishing Vessel Observation Network (FVON, providing wide spread high-resolution crowd sourced temperature data below the surface) on regional ocean model estimates in an eddy-dominated western boundary current region, the East Australian Current. In our companion paper, we used temporally and/or spatially averaged statistics to describe model-observation differences and show that SWOT substantially improves sea surface height representation across scales and FVON data strongly improves subsurface temperature representation in coastal and shelf regions. While these averaged statistics are useful to demonstrate overall impact of various observation platforms, they often do not capture the detail of how data assimilation reproduces fine-scale dynamical features - precisely the features that are resolved by SWOT and FVON. Here we present select case studies to demonstrate the specific time and space structure of observation impacts. Using independent cruise, glider, drifter and FVON observations, we show that SWOT and FVON can provide significant improvement in the representation of complex fine-scale oceanic structure. The results demonstrate that wide-swath altimetry and fishing-vessel subsurface observations provide complementary constraints: SWOT data improves the surface expression and evolution of eddies, while FVON data constrains the vertical thermal structure that surface observations alone cannot resolve. The case studies also reveal that small errors in feature placement can lead to large subsurface errors, underscoring the need for sustained subsurface observing and uncertainty-aware state estimation. Our results demonstrate that integrating SWOT and FVON observations into regional prediction systems offers a pathway toward more reliable ocean state estimates (essential for navigation, fisheries, marine heatwave monitoring, ecosystem applications, and coastal management) in an energetic boundary current system.