I welcome feedback on which scenario to use:
It’s a VERY simple model
Each FACT Table is completely Independent of other fact tables
I would appreciate your feedback in this format:
One Second Answer (the one you would use 80% of the time in your business): Scenario 2
Long Answer (includes reasons and best practice insights):
Quick answer - you will use two of these methods (Power Query and Dataset) and perhaps all three.
Long Answer is more of an explanation:
Power Query - the ‘shaping’ or transformation of your data using M code, this is used in both Datasets (which are multiple queries brought together into a single model), and in Dataflows
Dataset - this is what we call the entire data model, with multiple tables pulled together for the report
Dataflow - this is really Power Query that happens out in the PowerBi service, but it allows you to have a refresh on the individual tables that you are bringing together into the Dataset.
If you mean something different with your use of “Power Query”, “Dataset”, and “Dataflow”; please let me know so I can comment again 
Here are some resources that might help to better understand this:
Thank you for taking the time to provide feedback. I have watched the RADACAD several times, it was very helpful. I have updated our model based on your feedback and discussions in house to the following.
Happy to be of help 
