Add date-time related regressors

This question may already been asked but I did not find it and I' eager to have an answer.

I have a series with daily timestamp, such as:

```
2020-05-04 | 15
2020-05-05 | 3
2020-05-06 | 36
2020-05-07 | 11
2020-05-08 | 6
...
```

I know that Prophet takes care of date-time related features internally. However, is it reasonable to add day of the month and month of the year as an independent features to help Prophet to capture the trend and seasonality better? Such as:

```
ds | y | month | day
-----------------------------
2020-05-04 | 15 | 5 | 4
2020-05-05 | 3 | 5 | 5
2020-05-06 | 36 | 5 | 6
2020-05-07 | 11 | 5 | 7
2020-05-08 | 6 | 5 | 8
...
```

If I can do that, will this be similar to weighting each observation? I mean what will be the intuition behind including above mentioned two external regressors?