SysML difference in usage between object node and action pins

I've only recently started working with Cameo and I'm not sure exactly what the difference is between using object nodes and action pins.

My goal is to maintain consistency in my model where I want to use outputs from an activity or activity diagrams as input in the subsequent function/activity. I have the option of naming the input and output of two activities the same (although I don't expect the object flow to change) or creating an object node that connects the activities without action pins (whereby Cameo always automatically creates a new input pin, which makes me a little unsure about this approach).

Option 1

Option 2

Same problem when passing to another diagram. In the example here, I currently name all objects the same, but have no relationship in the model itself that indicates the same object apart from the name.

Forwarding to another diagram

Can someone explain to me the exact difference between these two variants and what is the common approach for such modeling?

Thanks in advance!

Jim is right. Option 2 is invalid. However, for some reasons, many people prefer to show an "object" in between two actions. This is allowed, but... it is a notational option that maps to the same model as option 1. So, one rectangle connected with arrows to two actions maps in the model to two pins and one object flow. I don't know any tool that supports this, though. Of course, you can use a drawing tool. Just make sure that you don't mix the "object"-notation with the pin-notation.