[barrkel]

There is very little difference (read: almost none) between a single thread of control transferring between objects and synchronous messages and associated return messages flowing between independent "microservices", providing the "microservices" aren't busy when not acting on a message or when waiting for the return value immediately after having sent a message.

In so far as things are easier to test in isolation, it'll be due to the more dynamically typed nature of message passing over and above traditional method calls.

"Microservices", or actors or whatever you want to call them, are more justified when you have concurrency inherent in your problem and you want to avoid the trouble threads create.